1 files changed, 219 insertions, 194 deletions

diff --git a/mm/mmap.c b/mm/mmap.c
index 13678edaa22c..9a93b054148a 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -76,10 +76,10 @@ int mmap_rnd_compat_bits __read_mostly = CONFIG_ARCH_MMAP_RND_COMPAT_BITS;
 static bool ignore_rlimit_data;
 core_param(ignore_rlimit_data, ignore_rlimit_data, bool, 0644);
 
-static void unmap_region(struct mm_struct *mm, struct maple_tree *mt,
+static void unmap_region(struct mm_struct *mm, struct ma_state *mas,
 		struct vm_area_struct *vma, struct vm_area_struct *prev,
 		struct vm_area_struct *next, unsigned long start,
-		unsigned long end, bool mm_wr_locked);
+		unsigned long end, unsigned long tree_end, bool mm_wr_locked);
 
 static pgprot_t vm_pgprot_modify(pgprot_t oldprot, unsigned long vm_flags)
 {
@@ -154,18 +154,6 @@ static inline struct vm_area_struct *vma_prev_limit(struct vma_iterator *vmi,
 	return mas_prev(&vmi->mas, min);
 }
 
-static inline int vma_iter_clear_gfp(struct vma_iterator *vmi,
-			unsigned long start, unsigned long end, gfp_t gfp)
-{
-	vmi->mas.index = start;
-	vmi->mas.last = end - 1;
-	mas_store_gfp(&vmi->mas, NULL, gfp);
-	if (unlikely(mas_is_err(&vmi->mas)))
-		return -ENOMEM;
-
-	return 0;
-}
-
 /*
  * check_brk_limits() - Use platform specific check of range & verify mlock
  * limits.
@@ -182,7 +170,8 @@ static int check_brk_limits(unsigned long addr, unsigned long len)
 	if (IS_ERR_VALUE(mapped_addr))
 		return mapped_addr;
 
-	return mlock_future_check(current->mm, current->mm->def_flags, len);
+	return mlock_future_ok(current->mm, current->mm->def_flags, len)
+		? 0 : -EAGAIN;
 }
 static int do_brk_flags(struct vma_iterator *vmi, struct vm_area_struct *brkvma,
 		unsigned long addr, unsigned long request, unsigned long flags);
@@ -300,61 +289,40 @@ out:
 }
 
 #if defined(CONFIG_DEBUG_VM_MAPLE_TREE)
-extern void mt_validate(struct maple_tree *mt);
-extern void mt_dump(const struct maple_tree *mt);
-
-/* Validate the maple tree */
-static void validate_mm_mt(struct mm_struct *mm)
-{
-	struct maple_tree *mt = &mm->mm_mt;
-	struct vm_area_struct *vma_mt;
-
-	MA_STATE(mas, mt, 0, 0);
-
-	mt_validate(&mm->mm_mt);
-	mas_for_each(&mas, vma_mt, ULONG_MAX) {
-		if ((vma_mt->vm_start != mas.index) ||
-		    (vma_mt->vm_end - 1 != mas.last)) {
-			pr_emerg("issue in %s\n", current->comm);
-			dump_stack();
-			dump_vma(vma_mt);
-			pr_emerg("mt piv: %p %lu - %lu\n", vma_mt,
-				 mas.index, mas.last);
-			pr_emerg("mt vma: %p %lu - %lu\n", vma_mt,
-				 vma_mt->vm_start, vma_mt->vm_end);
-
-			mt_dump(mas.tree);
-			if (vma_mt->vm_end != mas.last + 1) {
-				pr_err("vma: %p vma_mt %lu-%lu\tmt %lu-%lu\n",
-						mm, vma_mt->vm_start, vma_mt->vm_end,
-						mas.index, mas.last);
-				mt_dump(mas.tree);
-			}
-			VM_BUG_ON_MM(vma_mt->vm_end != mas.last + 1, mm);
-			if (vma_mt->vm_start != mas.index) {
-				pr_err("vma: %p vma_mt %p %lu - %lu doesn't match\n",
-						mm, vma_mt, vma_mt->vm_start, vma_mt->vm_end);
-				mt_dump(mas.tree);
-			}
-			VM_BUG_ON_MM(vma_mt->vm_start != mas.index, mm);
-		}
-	}
-}
-
 static void validate_mm(struct mm_struct *mm)
 {
 	int bug = 0;
 	int i = 0;
 	struct vm_area_struct *vma;
-	MA_STATE(mas, &mm->mm_mt, 0, 0);
-
-	validate_mm_mt(mm);
+	VMA_ITERATOR(vmi, mm, 0);
 
-	mas_for_each(&mas, vma, ULONG_MAX) {
+	mt_validate(&mm->mm_mt);
+	for_each_vma(vmi, vma) {
 #ifdef CONFIG_DEBUG_VM_RB
 		struct anon_vma *anon_vma = vma->anon_vma;
 		struct anon_vma_chain *avc;
+#endif
+		unsigned long vmi_start, vmi_end;
+		bool warn = 0;
+
+		vmi_start = vma_iter_addr(&vmi);
+		vmi_end = vma_iter_end(&vmi);
+		if (VM_WARN_ON_ONCE_MM(vma->vm_end != vmi_end, mm))
+			warn = 1;
+
+		if (VM_WARN_ON_ONCE_MM(vma->vm_start != vmi_start, mm))
+			warn = 1;
 
+		if (warn) {
+			pr_emerg("issue in %s\n", current->comm);
+			dump_stack();
+			dump_vma(vma);
+			pr_emerg("tree range: %px start %lx end %lx\n", vma,
+				 vmi_start, vmi_end - 1);
+			vma_iter_dump_tree(&vmi);
+		}
+
+#ifdef CONFIG_DEBUG_VM_RB
 		if (anon_vma) {
 			anon_vma_lock_read(anon_vma);
 			list_for_each_entry(avc, &vma->anon_vma_chain, same_vma)
@@ -365,14 +333,13 @@ static void validate_mm(struct mm_struct *mm)
 		i++;
 	}
 	if (i != mm->map_count) {
-		pr_emerg("map_count %d mas_for_each %d\n", mm->map_count, i);
+		pr_emerg("map_count %d vma iterator %d\n", mm->map_count, i);
 		bug = 1;
 	}
 	VM_BUG_ON_MM(bug, mm);
 }
 
 #else /* !CONFIG_DEBUG_VM_MAPLE_TREE */
-#define validate_mm_mt(root) do { } while (0)
 #define validate_mm(mm) do { } while (0)
 #endif /* CONFIG_DEBUG_VM_MAPLE_TREE */
 
@@ -441,7 +408,8 @@ static int vma_link(struct mm_struct *mm, struct vm_area_struct *vma)
 	VMA_ITERATOR(vmi, mm, 0);
 	struct address_space *mapping = NULL;
 
-	if (vma_iter_prealloc(&vmi))
+	vma_iter_config(&vmi, vma->vm_start, vma->vm_end);
+	if (vma_iter_prealloc(&vmi, vma))
 		return -ENOMEM;
 
 	if (vma->vm_file) {
@@ -694,19 +662,16 @@ int vma_expand(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	/* Only handles expanding */
 	VM_WARN_ON(vma->vm_start < start || vma->vm_end > end);
 
-	if (vma_iter_prealloc(vmi))
+	/* Note: vma iterator must be pointing to 'start' */
+	vma_iter_config(vmi, start, end);
+	if (vma_iter_prealloc(vmi, vma))
 		goto nomem;
 
 	vma_prepare(&vp);
 	vma_adjust_trans_huge(vma, start, end, 0);
-	/* VMA iterator points to previous, so set to start if necessary */
-	if (vma_iter_addr(vmi) != start)
-		vma_iter_set(vmi, start);
-
 	vma->vm_start = start;
 	vma->vm_end = end;
 	vma->vm_pgoff = pgoff;
-	/* Note: mas must be pointing to the expanding VMA */
 	vma_iter_store(vmi, vma);
 
 	vma_complete(&vp, vmi, vma->vm_mm);
@@ -733,19 +698,19 @@ int vma_shrink(struct vma_iterator *vmi, struct vm_area_struct *vma,
 
 	WARN_ON((vma->vm_start != start) && (vma->vm_end != end));
 
-	if (vma_iter_prealloc(vmi))
+	if (vma->vm_start < start)
+		vma_iter_config(vmi, vma->vm_start, start);
+	else
+		vma_iter_config(vmi, end, vma->vm_end);
+
+	if (vma_iter_prealloc(vmi, NULL))
 		return -ENOMEM;
 
 	init_vma_prep(&vp, vma);
 	vma_prepare(&vp);
 	vma_adjust_trans_huge(vma, start, end, 0);
 
-	if (vma->vm_start < start)
-		vma_iter_clear(vmi, vma->vm_start, start);
-
-	if (vma->vm_end > end)
-		vma_iter_clear(vmi, end, vma->vm_end);
-
+	vma_iter_clear(vmi);
 	vma->vm_start = start;
 	vma->vm_end = end;
 	vma->vm_pgoff = pgoff;
@@ -1021,7 +986,17 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm,
 	if (err)
 		return NULL;
 
-	if (vma_iter_prealloc(vmi))
+	if (vma_start < vma->vm_start || vma_end > vma->vm_end)
+		vma_expanded = true;
+
+	if (vma_expanded) {
+		vma_iter_config(vmi, vma_start, vma_end);
+	} else {
+		vma_iter_config(vmi, adjust->vm_start + adj_start,
+				adjust->vm_end);
+	}
+
+	if (vma_iter_prealloc(vmi, vma))
 		return NULL;
 
 	init_multi_vma_prep(&vp, vma, adjust, remove, remove2);
@@ -1030,8 +1005,6 @@ struct vm_area_struct *vma_merge(struct vma_iterator *vmi, struct mm_struct *mm,
 
 	vma_prepare(&vp);
 	vma_adjust_trans_huge(vma, vma_start, vma_end, adj_start);
-	if (vma_start < vma->vm_start || vma_end > vma->vm_end)
-		vma_expanded = true;
 
 	vma->vm_start = vma_start;
 	vma->vm_end = vma_end;
@@ -1167,21 +1140,21 @@ static inline unsigned long round_hint_to_min(unsigned long hint)
 	return hint;
 }
 
-int mlock_future_check(struct mm_struct *mm, unsigned long flags,
-		       unsigned long len)
+bool mlock_future_ok(struct mm_struct *mm, unsigned long flags,
+			unsigned long bytes)
 {
-	unsigned long locked, lock_limit;
+	unsigned long locked_pages, limit_pages;
 
-	/*  mlock MCL_FUTURE? */
-	if (flags & VM_LOCKED) {
-		locked = len >> PAGE_SHIFT;
-		locked += mm->locked_vm;
-		lock_limit = rlimit(RLIMIT_MEMLOCK);
-		lock_limit >>= PAGE_SHIFT;
-		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
-			return -EAGAIN;
-	}
-	return 0;
+	if (!(flags & VM_LOCKED) || capable(CAP_IPC_LOCK))
+		return true;
+
+	locked_pages = bytes >> PAGE_SHIFT;
+	locked_pages += mm->locked_vm;
+
+	limit_pages = rlimit(RLIMIT_MEMLOCK);
+	limit_pages >>= PAGE_SHIFT;
+
+	return locked_pages <= limit_pages;
 }
 
 static inline u64 file_mmap_size_max(struct file *file, struct inode *inode)
@@ -1293,7 +1266,7 @@ unsigned long do_mmap(struct file *file, unsigned long addr,
 		if (!can_do_mlock())
 			return -EPERM;
 
-	if (mlock_future_check(mm, vm_flags, len))
+	if (!mlock_future_ok(mm, vm_flags, len))
 		return -EAGAIN;
 
 	if (file) {
@@ -1475,6 +1448,48 @@ SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg)
 }
 #endif /* __ARCH_WANT_SYS_OLD_MMAP */
 
+static bool vm_ops_needs_writenotify(const struct vm_operations_struct *vm_ops)
+{
+	return vm_ops && (vm_ops->page_mkwrite || vm_ops->pfn_mkwrite);
+}
+
+static bool vma_is_shared_writable(struct vm_area_struct *vma)
+{
+	return (vma->vm_flags & (VM_WRITE | VM_SHARED)) ==
+		(VM_WRITE | VM_SHARED);
+}
+
+static bool vma_fs_can_writeback(struct vm_area_struct *vma)
+{
+	/* No managed pages to writeback. */
+	if (vma->vm_flags & VM_PFNMAP)
+		return false;
+
+	return vma->vm_file && vma->vm_file->f_mapping &&
+		mapping_can_writeback(vma->vm_file->f_mapping);
+}
+
+/*
+ * Does this VMA require the underlying folios to have their dirty state
+ * tracked?
+ */
+bool vma_needs_dirty_tracking(struct vm_area_struct *vma)
+{
+	/* Only shared, writable VMAs require dirty tracking. */
+	if (!vma_is_shared_writable(vma))
+		return false;
+
+	/* Does the filesystem need to be notified? */
+	if (vm_ops_needs_writenotify(vma->vm_ops))
+		return true;
+
+	/*
+	 * Even if the filesystem doesn't indicate a need for writenotify, if it
+	 * can writeback, dirty tracking is still required.
+	 */
+	return vma_fs_can_writeback(vma);
+}
+
 /*
  * Some shared mappings will want the pages marked read-only
  * to track write events. If so, we'll downgrade vm_page_prot
@@ -1483,21 +1498,18 @@ SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg)
  */
 int vma_wants_writenotify(struct vm_area_struct *vma, pgprot_t vm_page_prot)
 {
-	vm_flags_t vm_flags = vma->vm_flags;
-	const struct vm_operations_struct *vm_ops = vma->vm_ops;
-
 	/* If it was private or non-writable, the write bit is already clear */
-	if ((vm_flags & (VM_WRITE|VM_SHARED)) != ((VM_WRITE|VM_SHARED)))
+	if (!vma_is_shared_writable(vma))
 		return 0;
 
 	/* The backer wishes to know when pages are first written to? */
-	if (vm_ops && (vm_ops->page_mkwrite || vm_ops->pfn_mkwrite))
+	if (vm_ops_needs_writenotify(vma->vm_ops))
 		return 1;
 
 	/* The open routine did something to the protections that pgprot_modify
 	 * won't preserve? */
 	if (pgprot_val(vm_page_prot) !=
-	    pgprot_val(vm_pgprot_modify(vm_page_prot, vm_flags)))
+	    pgprot_val(vm_pgprot_modify(vm_page_prot, vma->vm_flags)))
 		return 0;
 
 	/*
@@ -1511,13 +1523,8 @@ int vma_wants_writenotify(struct vm_area_struct *vma, pgprot_t vm_page_prot)
 	if (userfaultfd_wp(vma))
 		return 1;
 
-	/* Specialty mapping? */
-	if (vm_flags & VM_PFNMAP)
-		return 0;
-
 	/* Can the mapping track the dirty pages? */
-	return vma->vm_file && vma->vm_file->f_mapping &&
-		mapping_can_writeback(vma->vm_file->f_mapping);
+	return vma_fs_can_writeback(vma);
 }
 
 /*
@@ -1911,7 +1918,7 @@ static int acct_stack_growth(struct vm_area_struct *vma,
 		return -ENOMEM;
 
 	/* mlock limit tests */
-	if (mlock_future_check(mm, vma->vm_flags, grow << PAGE_SHIFT))
+	if (!mlock_future_ok(mm, vma->vm_flags, grow << PAGE_SHIFT))
 		return -ENOMEM;
 
 	/* Check to ensure the stack will not grow into a hugetlb-only region */
@@ -1941,7 +1948,7 @@ int expand_upwards(struct vm_area_struct *vma, unsigned long address)
 	struct vm_area_struct *next;
 	unsigned long gap_addr;
 	int error = 0;
-	MA_STATE(mas, &mm->mm_mt, 0, 0);
+	MA_STATE(mas, &mm->mm_mt, vma->vm_start, address);
 
 	if (!(vma->vm_flags & VM_GROWSUP))
 		return -EFAULT;
@@ -1966,7 +1973,11 @@ int expand_upwards(struct vm_area_struct *vma, unsigned long address)
 		/* Check that both stack segments have the same anon_vma? */
 	}
 
-	if (mas_preallocate(&mas, GFP_KERNEL))
+	if (next)
+		mas_prev_range(&mas, address);
+
+	__mas_set_range(&mas, vma->vm_start, address - 1);
+	if (mas_preallocate(&mas, vma, GFP_KERNEL))
 		return -ENOMEM;
 
 	/* We must make sure the anon_vma is allocated. */
@@ -2009,7 +2020,6 @@ int expand_upwards(struct vm_area_struct *vma, unsigned long address)
 				anon_vma_interval_tree_pre_update_vma(vma);
 				vma->vm_end = address;
 				/* Overwrite old entry in mtree. */
-				mas_set_range(&mas, vma->vm_start, address - 1);
 				mas_store_prealloc(&mas, vma);
 				anon_vma_interval_tree_post_update_vma(vma);
 				spin_unlock(&mm->page_table_lock);
@@ -2048,7 +2058,11 @@ int expand_downwards(struct vm_area_struct *vma, unsigned long address)
 			return -ENOMEM;
 	}
 
-	if (mas_preallocate(&mas, GFP_KERNEL))
+	if (prev)
+		mas_next_range(&mas, vma->vm_start);
+
+	__mas_set_range(&mas, address, vma->vm_end - 1);
+	if (mas_preallocate(&mas, vma, GFP_KERNEL))
 		return -ENOMEM;
 
 	/* We must make sure the anon_vma is allocated. */
@@ -2092,7 +2106,6 @@ int expand_downwards(struct vm_area_struct *vma, unsigned long address)
 				vma->vm_start = address;
 				vma->vm_pgoff -= grow;
 				/* Overwrite old entry in mtree. */
-				mas_set_range(&mas, address, vma->vm_end - 1);
 				mas_store_prealloc(&mas, vma);
 				anon_vma_interval_tree_post_update_vma(vma);
 				spin_unlock(&mm->page_table_lock);
@@ -2205,18 +2218,20 @@ static inline void remove_mt(struct mm_struct *mm, struct ma_state *mas)
  *
  * Called with the mm semaphore held.
  */
-static void unmap_region(struct mm_struct *mm, struct maple_tree *mt,
+static void unmap_region(struct mm_struct *mm, struct ma_state *mas,
 		struct vm_area_struct *vma, struct vm_area_struct *prev,
-		struct vm_area_struct *next,
-		unsigned long start, unsigned long end, bool mm_wr_locked)
+		struct vm_area_struct *next, unsigned long start,
+		unsigned long end, unsigned long tree_end, bool mm_wr_locked)
 {
 	struct mmu_gather tlb;
+	unsigned long mt_start = mas->index;
 
 	lru_add_drain();
 	tlb_gather_mmu(&tlb, mm);
 	update_hiwater_rss(mm);
-	unmap_vmas(&tlb, mt, vma, start, end, mm_wr_locked);
-	free_pgtables(&tlb, mt, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS,
+	unmap_vmas(&tlb, mas, vma, start, end, tree_end, mm_wr_locked);
+	mas_set(mas, mt_start);
+	free_pgtables(&tlb, mas, vma, prev ? prev->vm_end : FIRST_USER_ADDRESS,
 				 next ? next->vm_start : USER_PGTABLES_CEILING,
 				 mm_wr_locked);
 	tlb_finish_mmu(&tlb);
@@ -2234,7 +2249,7 @@ int __split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	struct vm_area_struct *new;
 	int err;
 
-	validate_mm_mt(vma->vm_mm);
+	validate_mm(vma->vm_mm);
 
 	WARN_ON(vma->vm_start >= addr);
 	WARN_ON(vma->vm_end <= addr);
@@ -2249,10 +2264,6 @@ int __split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	if (!new)
 		return -ENOMEM;
 
-	err = -ENOMEM;
-	if (vma_iter_prealloc(vmi))
-		goto out_free_vma;
-
 	if (new_below) {
 		new->vm_end = addr;
 	} else {
@@ -2260,6 +2271,11 @@ int __split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
 		new->vm_pgoff += ((addr - vma->vm_start) >> PAGE_SHIFT);
 	}
 
+	err = -ENOMEM;
+	vma_iter_config(vmi, new->vm_start, new->vm_end);
+	if (vma_iter_prealloc(vmi, new))
+		goto out_free_vma;
+
 	err = vma_dup_policy(vma, new);
 	if (err)
 		goto out_free_vmi;
@@ -2292,7 +2308,7 @@ int __split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	/* Success. */
 	if (new_below)
 		vma_next(vmi);
-	validate_mm_mt(vma->vm_mm);
+	validate_mm(vma->vm_mm);
 	return 0;
 
 out_free_mpol:
@@ -2301,7 +2317,7 @@ out_free_vmi:
 	vma_iter_free(vmi);
 out_free_vma:
 	vm_area_free(new);
-	validate_mm_mt(vma->vm_mm);
+	validate_mm(vma->vm_mm);
 	return err;
 }
 
@@ -2318,21 +2334,6 @@ int split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	return __split_vma(vmi, vma, addr, new_below);
 }
 
-static inline int munmap_sidetree(struct vm_area_struct *vma,
-				   struct ma_state *mas_detach)
-{
-	vma_start_write(vma);
-	mas_set_range(mas_detach, vma->vm_start, vma->vm_end - 1);
-	if (mas_store_gfp(mas_detach, vma, GFP_KERNEL))
-		return -ENOMEM;
-
-	vma_mark_detached(vma, true);
-	if (vma->vm_flags & VM_LOCKED)
-		vma->vm_mm->locked_vm -= vma_pages(vma);
-
-	return 0;
-}
-
 /*
  * do_vmi_align_munmap() - munmap the aligned region from @start to @end.
  * @vmi: The vma iterator
@@ -2354,6 +2355,7 @@ do_vmi_align_munmap(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	struct maple_tree mt_detach;
 	int count = 0;
 	int error = -ENOMEM;
+	unsigned long locked_vm = 0;
 	MA_STATE(mas_detach, &mt_detach, 0, 0);
 	mt_init_flags(&mt_detach, vmi->mas.tree->ma_flags & MT_FLAGS_LOCK_MASK);
 	mt_set_external_lock(&mt_detach, &mm->mmap_lock);
@@ -2377,81 +2379,85 @@ do_vmi_align_munmap(struct vma_iterator *vmi, struct vm_area_struct *vma,
 		if (end < vma->vm_end && mm->map_count >= sysctl_max_map_count)
 			goto map_count_exceeded;
 
-		error = __split_vma(vmi, vma, start, 0);
+		error = __split_vma(vmi, vma, start, 1);
 		if (error)
 			goto start_split_failed;
-
-		vma = vma_iter_load(vmi);
 	}
 
-	prev = vma_prev(vmi);
-	if (unlikely((!prev)))
-		vma_iter_set(vmi, start);
-
 	/*
 	 * Detach a range of VMAs from the mm. Using next as a temp variable as
 	 * it is always overwritten.
 	 */
-	for_each_vma_range(*vmi, next, end) {
+	next = vma;
+	do {
 		/* Does it split the end? */
 		if (next->vm_end > end) {
 			error = __split_vma(vmi, next, end, 0);
 			if (error)
 				goto end_split_failed;
 		}
-		error = munmap_sidetree(next, &mas_detach);
-		if (error)
-			goto munmap_sidetree_failed;
+		vma_start_write(next);
+		if (mas_store_gfp(&mas_detach, next, GFP_KERNEL))
+			goto munmap_gather_failed;
+		vma_mark_detached(next, true);
+		if (next->vm_flags & VM_LOCKED)
+			locked_vm += vma_pages(next);
 
 		count++;
+		if (unlikely(uf)) {
+			/*
+			 * If userfaultfd_unmap_prep returns an error the vmas
+			 * will remain split, but userland will get a
+			 * highly unexpected error anyway. This is no
+			 * different than the case where the first of the two
+			 * __split_vma fails, but we don't undo the first
+			 * split, despite we could. This is unlikely enough
+			 * failure that it's not worth optimizing it for.
+			 */
+			error = userfaultfd_unmap_prep(next, start, end, uf);
+
+			if (error)
+				goto userfaultfd_error;
+		}
 #ifdef CONFIG_DEBUG_VM_MAPLE_TREE
 		BUG_ON(next->vm_start < start);
 		BUG_ON(next->vm_start > end);
 #endif
-	}
-
-	next = vma_next(vmi);
-	if (unlikely(uf)) {
-		/*
-		 * If userfaultfd_unmap_prep returns an error the vmas
-		 * will remain split, but userland will get a
-		 * highly unexpected error anyway. This is no
-		 * different than the case where the first of the two
-		 * __split_vma fails, but we don't undo the first
-		 * split, despite we could. This is unlikely enough
-		 * failure that it's not worth optimizing it for.
-		 */
-		error = userfaultfd_unmap_prep(mm, start, end, uf);
-
-		if (error)
-			goto userfaultfd_error;
-	}
+	} for_each_vma_range(*vmi, next, end);
 
 #if defined(CONFIG_DEBUG_VM_MAPLE_TREE)
 	/* Make sure no VMAs are about to be lost. */
 	{
-		MA_STATE(test, &mt_detach, start, end - 1);
+		MA_STATE(test, &mt_detach, 0, 0);
 		struct vm_area_struct *vma_mas, *vma_test;
 		int test_count = 0;
 
 		vma_iter_set(vmi, start);
 		rcu_read_lock();
-		vma_test = mas_find(&test, end - 1);
+		vma_test = mas_find(&test, count - 1);
 		for_each_vma_range(*vmi, vma_mas, end) {
 			BUG_ON(vma_mas != vma_test);
 			test_count++;
-			vma_test = mas_next(&test, end - 1);
+			vma_test = mas_next(&test, count - 1);
 		}
 		rcu_read_unlock();
 		BUG_ON(count != test_count);
 	}
 #endif
-	/* Point of no return */
-	vma_iter_set(vmi, start);
+	error = -ENOMEM;
+	while (vma_iter_addr(vmi) > start)
+		vma_iter_prev_range(vmi);
+
 	if (vma_iter_clear_gfp(vmi, start, end, GFP_KERNEL))
-		return -ENOMEM;
+		goto clear_tree_failed;
 
+	mm->locked_vm -= locked_vm;
 	mm->map_count -= count;
+	prev = vma_iter_prev_range(vmi);
+	next = vma_next(vmi);
+	if (next)
+		vma_iter_prev_range(vmi);
+
 	/*
 	 * Do not downgrade mmap_lock if we are next to VM_GROWSDOWN or
 	 * VM_GROWSUP VMA. Such VMAs can change their size under
@@ -2470,9 +2476,11 @@ do_vmi_align_munmap(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	 * We can free page tables without write-locking mmap_lock because VMAs
 	 * were isolated before we downgraded mmap_lock.
 	 */
-	unmap_region(mm, &mt_detach, vma, prev, next, start, end, !downgrade);
+	mas_set(&mas_detach, 1);
+	unmap_region(mm, &mas_detach, vma, prev, next, start, end, count,
+		     !downgrade);
 	/* Statistics and freeing VMAs */
-	mas_set(&mas_detach, start);
+	mas_set(&mas_detach, 0);
 	remove_mt(mm, &mas_detach);
 	__mt_destroy(&mt_detach);
 
@@ -2480,9 +2488,14 @@ do_vmi_align_munmap(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	validate_mm(mm);
 	return downgrade ? 1 : 0;
 
+clear_tree_failed:
 userfaultfd_error:
-munmap_sidetree_failed:
+munmap_gather_failed:
 end_split_failed:
+	mas_set(&mas_detach, 0);
+	mas_for_each(&mas_detach, next, end)
+		vma_mark_detached(next, false);
+
 	__mt_destroy(&mt_detach);
 start_split_failed:
 map_count_exceeded:
@@ -2590,8 +2603,12 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
 
 	next = vma_next(&vmi);
 	prev = vma_prev(&vmi);
-	if (vm_flags & VM_SPECIAL)
+	if (vm_flags & VM_SPECIAL) {
+		if (prev)
+			vma_iter_next_range(&vmi);
+
 		goto cannot_expand;
+	}
 
 	/* Attempt to expand an old mapping */
 	/* Check next */
@@ -2601,6 +2618,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
 		merge_end = next->vm_end;
 		vma = next;
 		vm_pgoff = next->vm_pgoff - pglen;
+
 	}
 
 	/* Check prev */
@@ -2612,9 +2630,10 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
 		merge_start = prev->vm_start;
 		vma = prev;
 		vm_pgoff = prev->vm_pgoff;
+	} else if (prev) {
+		vma_iter_next_range(&vmi);
 	}
 
-
 	/* Actually expand, if possible */
 	if (vma &&
 	    !vma_expand(&vmi, vma, merge_start, merge_end, vm_pgoff, next)) {
@@ -2634,7 +2653,7 @@ cannot_expand:
 		goto unacct_error;
 	}
 
-	vma_iter_set(&vmi, addr);
+	vma_iter_config(&vmi, addr, end);
 	vma->vm_start = addr;
 	vma->vm_end = end;
 	vm_flags_init(vma, vm_flags);
@@ -2661,7 +2680,7 @@ cannot_expand:
 		if (WARN_ON((addr != vma->vm_start)))
 			goto close_and_free_vma;
 
-		vma_iter_set(&vmi, addr);
+		vma_iter_config(&vmi, addr, end);
 		/*
 		 * If vm_flags changed after call_mmap(), we should try merge
 		 * vma again as we may succeed this time.
@@ -2708,7 +2727,7 @@ cannot_expand:
 		goto close_and_free_vma;
 
 	error = -ENOMEM;
-	if (vma_iter_prealloc(&vmi))
+	if (vma_iter_prealloc(&vmi, vma))
 		goto close_and_free_vma;
 
 	if (vma->vm_file)
@@ -2777,9 +2796,10 @@ unmap_and_free_vma:
 		fput(vma->vm_file);
 		vma->vm_file = NULL;
 
+		vma_iter_set(&vmi, vma->vm_end);
 		/* Undo any partial mapping done by a device driver. */
-		unmap_region(mm, &mm->mm_mt, vma, prev, next, vma->vm_start,
-			     vma->vm_end, true);
+		unmap_region(mm, &vmi.mas, vma, prev, next, vma->vm_start,
+			     vma->vm_end, vma->vm_end, true);
 	}
 	if (file && (vm_flags & VM_SHARED))
 		mapping_unmap_writable(file->f_mapping);
@@ -2936,7 +2956,7 @@ int do_vma_munmap(struct vma_iterator *vmi, struct vm_area_struct *vma,
 
 	arch_unmap(mm, start, end);
 	ret = do_vmi_align_munmap(vmi, vma, mm, start, end, uf, downgrade);
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	return ret;
 }
 
@@ -2958,7 +2978,7 @@ static int do_brk_flags(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	struct mm_struct *mm = current->mm;
 	struct vma_prepare vp;
 
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	/*
 	 * Check against address space limits by the changed size
 	 * Note: This happens *after* clearing old mappings in some code paths.
@@ -2980,7 +3000,8 @@ static int do_brk_flags(struct vma_iterator *vmi, struct vm_area_struct *vma,
 	if (vma && vma->vm_end == addr && !vma_policy(vma) &&
 	    can_vma_merge_after(vma, flags, NULL, NULL,
 				addr >> PAGE_SHIFT, NULL_VM_UFFD_CTX, NULL)) {
-		if (vma_iter_prealloc(vmi))
+		vma_iter_config(vmi, vma->vm_start, addr + len);
+		if (vma_iter_prealloc(vmi, vma))
 			goto unacct_fail;
 
 		init_vma_prep(&vp, vma);
@@ -2995,6 +3016,8 @@ static int do_brk_flags(struct vma_iterator *vmi, struct vm_area_struct *vma,
 		goto out;
 	}
 
+	if (vma)
+		vma_iter_next_range(vmi);
 	/* create a vma struct for an anonymous mapping */
 	vma = vm_area_alloc(mm);
 	if (!vma)
@@ -3108,7 +3131,7 @@ void exit_mmap(struct mm_struct *mm)
 	tlb_gather_mmu_fullmm(&tlb, mm);
 	/* update_hiwater_rss(mm) here? but nobody should be looking */
 	/* Use ULONG_MAX here to ensure all VMAs in the mm are unmapped */
-	unmap_vmas(&tlb, &mm->mm_mt, vma, 0, ULONG_MAX, false);
+	unmap_vmas(&tlb, &mas, vma, 0, ULONG_MAX, ULONG_MAX, false);
 	mmap_read_unlock(mm);
 
 	/*
@@ -3118,7 +3141,8 @@ void exit_mmap(struct mm_struct *mm)
 	set_bit(MMF_OOM_SKIP, &mm->flags);
 	mmap_write_lock(mm);
 	mt_clear_in_rcu(&mm->mm_mt);
-	free_pgtables(&tlb, &mm->mm_mt, vma, FIRST_USER_ADDRESS,
+	mas_set(&mas, vma->vm_end);
+	free_pgtables(&tlb, &mas, vma, FIRST_USER_ADDRESS,
 		      USER_PGTABLES_CEILING, true);
 	tlb_finish_mmu(&tlb);
 
@@ -3127,6 +3151,7 @@ void exit_mmap(struct mm_struct *mm)
 	 * enabled, without holding any MM locks besides the unreachable
 	 * mmap_write_lock.
 	 */
+	mas_set(&mas, vma->vm_end);
 	do {
 		if (vma->vm_flags & VM_ACCOUNT)
 			nr_accounted += vma_pages(vma);
@@ -3199,7 +3224,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
 	bool faulted_in_anon_vma = true;
 	VMA_ITERATOR(vmi, mm, addr);
 
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	/*
 	 * If anonymous vma has not yet been faulted, update new pgoff
 	 * to match new location, to increase its chance of merging.
@@ -3258,7 +3283,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
 			goto out_vma_link;
 		*need_rmap_locks = false;
 	}
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	return new_vma;
 
 out_vma_link:
@@ -3274,7 +3299,7 @@ out_free_mempol:
 out_free_vma:
 	vm_area_free(new_vma);
 out:
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	return NULL;
 }
 
@@ -3411,7 +3436,7 @@ static struct vm_area_struct *__install_special_mapping(
 	int ret;
 	struct vm_area_struct *vma;
 
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	vma = vm_area_alloc(mm);
 	if (unlikely(vma == NULL))
 		return ERR_PTR(-ENOMEM);
@@ -3434,12 +3459,12 @@ static struct vm_area_struct *__install_special_mapping(
 
 	perf_event_mmap(vma);
 
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	return vma;
 
 out:
 	vm_area_free(vma);
-	validate_mm_mt(mm);
+	validate_mm(mm);
 	return ERR_PTR(ret);
 }